Intelligent identification method for strip shape regulation and control efficacy based on data driving

Abstract

The invention relates to the technical field of cold-rolled strip steel, in particular to an intelligent strip shape regulation and control efficacy identification method based on data driving, which comprises the following steps: acquiring industrial data in the production process of a rolling mill unit, preprocessing the industrial data, and establishing a data set based on the preprocessed industrial data, the industrial data comprises process record data and plate shape instrument data; and performing noise reduction processing, decomposition and reconstruction on the strip shape instrument data in the data set, and performing intelligent identification of regulation and control efficacy on the data set after noise reduction processing, decomposition and reconstruction by using an adaptive moment estimation algorithm. According to the method, the regulation and control capability of the regulation and control mechanism can be analyzed and determined, and more control information is provided for subsequent strip shape control calculation regulation and control quantity.

Description

Technical Field

[0001] The invention relates to the technical field of cold-rolled strip steel, and in particular to a data-driven intelligent identification method for plate shape control efficacy. Background Art

[0002] The steel industry is an important part of my country's industry. The development level of the steel industry can represent the industrialization level of a country to a certain extent. Cold-rolled strip has the characteristics of high output, fast forming speed and excellent performance, and is widely used in automobile manufacturing, electrical products and other industries. With the improvement of the manufacturing level, all walks of life have higher and higher requirements for the precision of strip materials. The quality indicators of strip materials mainly include thickness and shape. At present, the transverse thickness control has basically met the requirements, and there are still some unresolved problems in shape control. Improving the accuracy of shape control is an important part of improving the quality of shape.

[0003] The fifth stand of the six-high CVC mill uses a closed-loop feedback control strategy to fine-tune the plate shape, which is the most critical part of the plate shape control system. The plate shape closed-loop feedback control system is established in the stable rolling stage. The deviation between the measured plate shape returned by the plate shape meter and the target plate shape is calculated. The closed-loop feedback control model calculates the control amount of the actuator according to the control efficiency of each actuator, sends a control signal to the actuator, and detects new plate shape deviations at the same time. The actuator is continuously adjusted in real time and dynamically, and finally a stable and good plate shape is obtained. At present, the control mechanisms of the five-stand cold rolling mill include intermediate roll bending (IRB), work roll bending (WRB), intermediate roll shifting (IRS), tilting and segmented cooling. Among them, tilting can eliminate the first and third order plate shape deviations, bending and shifting can eliminate the second and fourth order plate shape deviations, and higher order plate shape deviations need to be eliminated by segmented cooling. The control performance of each actuator can be reflected on the control efficiency curve. Generally, segmented cooling cannot be solved by formulas and has a slow response speed. It is often used to eliminate plate shape deviations that cannot be eliminated by other actuators. Therefore, only the effects of pressing down tilt, bending roll and shifting roll are considered here.

[0004] The control effect is the basis of plate shape control, which can effectively reflect the control performance of the actuator on the plate shape. Because its calculation process is simple and fast, it can play a role in the unit with multiple control means. Therefore, it can make full use of the rolling production process data to optimize it, so as to obtain more accurate plate shape control. The control effect will affect the accuracy of plate shape control. In actual production, the control effect of each actuator varies with different process parameters, especially when the width-to-thickness ratio is large, it is more unstable. However, the control effect used on site is often used for a variety of working conditions, there are certain errors for specific working conditions, and the update speed is slow. Therefore, obtaining accurate control effect is the focus of plate shape control research. Summary of the invention

[0005] The purpose of the present invention is to provide a data-driven intelligent identification method for plate shape control efficacy, clarify the control capabilities of various control mechanisms, solve the problem of fixed control efficacy, improve the accuracy of plate shape control, and realize intelligent identification of control efficacy.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A data-driven intelligent identification method for plate shape control efficacy, comprising:

[0008] Acquire industrial data in the production process of the rolling mill unit, preprocess the industrial data, and establish a data set based on the preprocessed industrial data, wherein the industrial data includes process record data and flatness meter data;

[0009] The shape meter data in the data set is subjected to denoising and decomposition and reconstruction, and an adaptive moment estimation algorithm is used to intelligently identify the control efficacy of the denoising and decomposition and reconstruction data set.

[0010] Furthermore, preprocessing the industrial data includes: performing outlier processing on the process record data.

[0011] Furthermore, establishing a data set based on the preprocessed industrial data includes:

[0012] Based on the exit strip length, the pre-processed process record data and the shape meter data are matched in time series to construct the data set.

[0013] Furthermore, the process of performing the corresponding matching includes: respectively acquiring a first shape deviation in the process record data and a second shape deviation in the shape meter data, and performing the corresponding matching based on a comparative verification of the first shape deviation and the second shape deviation.

[0014] Further, performing noise reduction processing on the shape meter data in the data set includes:

[0015] The wavelet transform algorithm is used to perform noise reduction on the shape meter data, and the formula is:

[0016] s(n)=f(n)+σe(n)

[0017] Where s(n) is the actual noisy signal; f(n) is the effective signal; n is the sampling time; σ is the noise intensity; and e(n) is the noise component.

[0018] Further, using a wavelet transform algorithm to perform noise reduction processing on the shape meter data includes:

[0019] The shape meter data is subjected to multi-layer decomposition of the original signal, using max_level = log 2 (l data / l filter ) calculate the number of layers of the multi-layer decomposition; after determining the number of decomposition layers, optimize the threshold of the wavelet coefficient using a genetic algorithm; finally, reconstruct the signal using an inverse wavelet transform to obtain the denoised plate shape meter data.

[0020] Further, decomposing and reconstructing the shape meter data in the data set includes:

[0021] The shape meter data is decomposed based on Legendre orthogonal polynomials to obtain the coefficients λ of each shape basis mode: 1 ,λ 2 ,λ 3 and λ 4 ;

[0022] Based on the coefficients λ of each plate shape basis mode 1 ,λ 2 ,λ 3 and λ 4 The shape meter data is reconstructed to obtain decomposed and reconstructed shape meter data.

[0023] Furthermore, the adaptive moment estimation algorithm is used to intelligently identify the control effect of the data set after noise reduction and decomposition and reconstruction, including:

[0024] Based on the adaptive moment estimation algorithm, the learning rate is automatically updated according to the data characteristics of the data set using the first-order and second-order moment estimates of the gradient, and the first-order and second-order moment estimates are biased, and an updated calculation model for the variable x is defined based on the bias correction of the first-order and second-order moment estimates. The variable x is updated and calculated based on the updated calculation model of the variable x, thereby realizing intelligent identification of the regulation efficacy.

[0025] Furthermore, the update calculation model of the variable x is:

[0026]

[0027] In the formula, x t is the x variable at the current moment; t-1 is the x variable at the previous moment; η∈(0,1] is the learning rate; ε is a constant; is the bias correction for the first-order moment estimate; is the bias correction for the second-order moment estimate.

[0028] The beneficial effects of the present invention are:

[0029] Starting from the actual production data of the industrial site, the present invention performs data preprocessing operations such as outlier processing and data matching on the process record data and the shape meter data to establish a data set; performs noise reduction processing on the shape meter data, and uses the adaptive moment estimation algorithm to intelligently identify the control effect of the processed data set. The control ability of the control mechanism is analyzed and clarified, providing more control information for the subsequent shape control calculation and control quantity. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0031] Figure 1 A flow chart of a data-driven intelligent identification method for plate shape control efficacy according to an embodiment of the present invention;

[0032] Figure 2 A comparison diagram of the shape deviation between the process record data and the shape meter data of an embodiment of the present invention;

[0033] Figure 3 This is a diagram of the GA-WT denoising process according to an embodiment of the present invention;

[0034] Figure 4 1 is a comparison diagram of three denoising methods according to an embodiment of the present invention, wherein (a) is a denoising effect diagram of GA-WT, (b) is a denoising effect diagram of WT, and (c) is a denoising effect diagram of EEMD;

[0035] Figure 5 : is a diagram showing the basis modes of various plate shape defects of Legendre orthogonal polynomials according to an embodiment of the present invention, wherein Y 1 For the right wave, Y 2 For the left wave, Y 3 For double wave, Y 4 For the middle wave, Y 5 It is the left three-point wave, Y 6 It is the right three-point wave, Y 7 It is a side-middle composite wave, Y 8 It is a four-point wave;

[0036] Figure 6 This is a comparison chart of the control effects of the adaptive moment estimation algorithm recognition and on-site recognition according to an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0039] This embodiment provides a data-driven intelligent identification method for plate shape control efficacy. Figure 1 As shown, including:

[0040] S1. Obtain industrial data during the actual production process of the rolling mill unit and pre-process the industrial data;

[0041] Industrial data includes process record data and shape meter data. The shape meter data includes the strip residual stress deviation and strip length signal in each measuring area. The strip residual stress deviation is processed and converted into shape deviation. The process record data is processed for outlier value by data cleaning.

[0042] In the exported files of process record data, null values ​​often appear because the sampling time of sensors is not exactly the same or because of electrical failures, which leads to errors in numerical statistics and calculations. Therefore, null values ​​need to be filled. If the value before the null value is valid, its value will replace the null value; if the null value is in the first row, the first non-null value queried backward will be filled into the null value row.

[0043] S2, perform data matching and establish a data set;

[0044] To analyze the influence of the process parameters of the cold rolling mill on the plate shape, it is necessary to match the process parameters and the plate shape meter data in time series, that is, to match the process record data and the plate shape meter data one by one according to the time series.

[0045] The only signal in the process recording data that corresponds to the flatness is the length of the exit strip. Therefore, the exit strip length signal is used to match the process recording data and the flatness meter data. Then, the flatness deviation is obtained from the flatness meter data and compared with the flatness deviation signal in the process recording data to verify the degree of matching. If the degree of matching meets the preset degree, a data set is established.

[0046] Take the time series matching of the process record data and the shape meter data of a steel coil as an example. The matching standard is the length of the export strip, and the verification standard is the first shape deviation in the process record data and the second shape deviation in the shape meter data. The shape deviation verification diagram after splicing is as follows: Figure 2 As shown, the unit of flatness deviation is IU.

[0047] S3, using a wavelet transform algorithm based on a genetic algorithm to perform noise reduction on the shape meter data;

[0048] Wavelet transform is a time-frequency localized analysis method that solves the shortcoming of short-time Fourier transform that uses a fixed sliding window function and cannot change the time resolution. The area of ​​the wavelet window remains unchanged, but the window shape can change adaptively. When analyzing low-frequency signals, the time window of wavelet transform becomes longer to obtain more effective signals; when analyzing high-frequency signals, the window becomes shorter to refine the local signal.

[0049] Wavelet is a time function with a very short duration. When the Fourier transform of the integrable function ψ(t) satisfies the following equation, ψ(t) is called a basic wavelet, as shown in the following equation:

[0050]

[0051] The set of basic wavelets {ψ a,b (t)} can be obtained by scaling and translating the wavelet ψ(t), as shown in the following formula:

[0052]

[0053] In the formula, b is the translation factor, which determines the position of the function translation; a is the scale factor, which determines the scale of the function scaling.

[0054] For a signal f(t)∈L 2 (R), its continuous wavelet transform is shown as follows:

[0055]

[0056] Where W f (a, b) are the coefficients obtained by wavelet transform of signal f(t).

[0057] The inverse transform of f(t) can restore the signal as shown below:

[0058]

[0059] Discretize the wavelet and take the scale factor The translation factor is j is the power of the control scale factor. The larger j is, the smaller the scale is and the higher the frequency is. k is the position parameter of the control translation factor. a0 is the step value; is the change of wavelet function in scale; b 0 is the basic scale factor, which is used to determine the basic scale of the wavelet. The step value is a fixed value other than 1, usually a 0 >1, then It can be written as follows:

[0060]

[0061] The corresponding discrete wavelet transform can be written as follows:

[0062]

[0063] Wavelet transform can effectively identify signals of various frequencies, remove noise signals and retain the effective information and mutations of the signals. The formula for removing noise using wavelet transform is as follows:

[0064] s(n)=f(n)+σe(n)

[0065] Where s(n) is the actual noisy signal; f(n) is the effective signal; n is the sampling time; σ is the noise intensity; and e(n) is the noise component.

[0066] Using wavelet transform to remove noise requires multi-layer decomposition of the original signal, threshold processing of wavelet coefficients, reconstruction of inverse wavelet transform, etc. The more decomposition layers there are, the easier it is to remove the noise signal, but the error after reconstruction will be greater, so choosing the appropriate number of decomposition layers is crucial. The following formula can be used to calculate the appropriate number of decomposition layers:

[0067] max_level=log 2 (l data / l filter )

[0068] Where max_level is the number of decomposition levels; l data is the length of the data; l filter is the length of the filter.

[0069] The threshold processing of wavelet coefficients is the key to improving the denoising effect of wavelet transform, and mainly uses hard threshold function or soft threshold function. Since the soft threshold function can shrink the large coefficients, it can reduce the discontinuity problem in the hard threshold denoising process. At the same time, it has good adaptability and better denoising effect. Therefore, the soft threshold method is selected, and the expression is shown in the following formula:

[0070]

[0071] Where λ is the threshold; w is the original wavelet coefficient; w λare the new wavelet coefficients.

[0072] In order to measure the noise reduction performance, the signal-to-noise ratio (SNR) and root mean square error (RMSE) are generally used as evaluation criteria. The signal-to-noise ratio (SNR) is defined as the ratio of the effective component of the reconstructed signal to the noise after the noise is removed from the original signal. The larger the ratio, the better the noise reduction performance. The calculation formula is shown below:

[0073]

[0074] Where y(n) is the original signal; x(n) is the signal after noise reduction; N is the signal length.

[0075] The root mean square error (RMSE) reflects the similarity between the reconstructed signal and the original signal. The smaller the value, the closer it is to the original signal. Its definition is as follows:

[0076]

[0077] When denoising actual data, using wavelet transform alone may result in improper threshold selection, resulting in signal distortion or limited denoising effect. Therefore, the threshold of wavelet coefficients can be optimized in combination with genetic algorithm to remove noise as much as possible while retaining the characteristics of the signal. Selecting a suitable threshold is the key to wavelet denoising. In order to obtain the optimal wavelet coefficient, this embodiment optimizes the threshold in combination with genetic algorithm, and proposes a wavelet transform algorithm based on genetic algorithm (GA-WT). After wavelet transform decomposes to obtain wavelet coefficients, a suitable threshold is selected using genetic algorithm to remove noise in high frequencies and retain useful information. Then, the signal is reconstructed using inverse wavelet transform to obtain denoised data. Figure 3 is the denoising process of GA-WT, where cD j is the high-frequency wavelet coefficient obtained by decomposition, cA j is the low-frequency wavelet coefficient obtained by decomposition.

[0078] The commonly used EEMD (Ensemble Empirical Mode Decomposition) method and the wavelet transform method without genetic algorithm optimization are compared with the GA-WT method proposed in this embodiment, and SNR and RMSE are used as evaluation indicators.

[0079] To establish a GA-WT model, we first need to select a suitable wavelet basis function. In this embodiment, we select the db8 basis function, and max_level = log 2 (l data / l filter) The decomposition layer is calculated to be 6. The low-frequency wavelet coefficients are retained, and the genetic algorithm is used to select the optimal threshold for the high-frequency wavelet coefficients. The initial population of the genetic algorithm is 200, the crossover probability is 0.8, the number of iterations is 50, and the mutation probability is 0.003.

[0080] Figure 4 The following is a comparison of the denoising effects of the three methods. Figure (a) is the denoising effect of GA-WT, Figure (b) is the denoising effect of WT, and Figure (c) is the denoising effect of EEMD. It can be seen that the three methods can basically restore the signal trend, but the wavelet transform based on the genetic algorithm is better in details, and the curve transition is smoother. The unoptimized wavelet transform and EEMD have large fluctuations, do not remove more noise, and some effective information is removed. The denoising effects of the three methods are shown in Table 1. The SNR of GA-WT is 16.76db, which is higher than the 11.64db of the ordinary wavelet transform and the 13.46db of EEMD. At the same time, the RMSE of GA-WT is 0.51IU, which is lower than the 0.92IU of the ordinary wavelet transform and the 0.76IU of EEMD. It can be seen that the denoising performance of GA-WT is better than the other two methods, so the GA-WT method is selected to denoise the measured plate shape data.

[0081] Table 1

[0082]

[0083] S4, decomposing and reconstructing the shape meter data based on Legendre orthogonal polynomials;

[0084] The shape meter data after the noise reduction process in step S3 is decomposed based on the first wave shape basis mode, the second wave shape basis mode, the third wave shape basis mode, and the fourth wave shape basis mode using Legendre orthogonal polynomials. The first wave shape basis mode, the second wave shape basis mode, the third wave shape basis mode, and the fourth wave shape basis mode are respectively expressed as δ 1 ,δ 2 ,δ 3 ,δ 4 The formula is as follows: 1 (x) = x, After decomposition, the coefficients of each basis mode are obtained as λ 1 ,λ 2 ,λ 3 ,λ 4 .

[0085] like Figure 5 As shown in the figure, the common wave shape of plate shape defects is represented by the basis mode of Legendre orthogonal polynomial, where Y 1 For the right wave, Y 2 For the left wave, Y 3For double wave, Y 4 For the middle wave, Y 5 It is the left three-point wave, Y 6 It is the right three-point wave, Y 7 It is a side-middle composite wave, Y 8 It is a quarter wave. The basic patterns of common wave shapes include:

[0086] Right wave: Y 1 =δ 1 (x) = x;

[0087] Left wave: Y 2 =-δ 1 (x) = -x;

[0088] Double wave:

[0089] Middle Wave:

[0090] Left three-point wave:

[0091] Right three-point wave:

[0092] Edge-middle compound wave:

[0093] Quarter Wave:

[0094] Based on the coefficients λ of each basis mode obtained by decomposition 1 ,λ 2 ,λ 3 ,λ 4 , and then through Δflat=λ 1 δ 1 +λ 2 δ 2 +λ 3 δ 3 +λ 4 δ 4 Reconstruct the plate shape and realize filtering processing of the plate shape meter data.

[0095] S5. Using an adaptive moment estimation algorithm to intelligently identify the control effect of the control mechanism, and intuitively describe the control effect through curve graphics;

[0096] In this example, the objective function for defining the intelligent identification of plate shape control efficacy is as follows:

[0097]

[0098] In the formula, w j is the control value of the jth actuator; eff jiis the control efficiency of the jth actuator in the i-th section of the bandwidth direction; n is the number of flatness deviation values; Δflat i is the flatness deviation of the i-th section in the bandwidth direction.

[0099] In order to minimize the objective function J(eff), the gradient matrix g at the current moment is calculated by the following formula: t :

[0100]

[0101] The adaptive moment estimation algorithm is a variant of the gradient descent method, which makes up for the inability of the learning rate to change in the stochastic gradient descent method. The adaptive moment estimation algorithm uses the first-order and second-order moment estimates of the gradient to automatically update the learning rate according to the characteristics of the data. The current moment gradient calculated by the above formula is g t , then the first-order moment estimate M t As shown below:

[0102] M t =β 1 ·M t-1 +(1-β 1 )·g t

[0103] Where t is the current time; β 1 ∈[0,1) is the exponential decay rate of the moment, which is generally taken as 0.9; M t-1 is the first-order moment estimate of the previous moment; g t is the gradient at the current moment.

[0104] Second-order moment estimate V t As shown below:

[0105] V t =β 2 ·V t-1 +(1-β 2 )·g t 2

[0106] Where V t-1 is the second-order moment estimate of the previous moment; β 2 ∈[0,1) is the exponential decay rate of the second-order moment, which is generally taken as 0.999.

[0107] When the first-order and second-order moment estimates are initialized to zero vectors, the initialization step size will be too large, so they need to be corrected for deviations. Take the second-order moment as an example to calculate the deviation correction term. t Approaching 0, the second-order moment estimation formula can be expanded to obtain the following formula:

[0108]

[0109] In the formula, is the exponential decay rate of the second-order moment at the previous moment; g i 2 is the gradient at the i-th moment.

[0110] Taking the expectation on both sides of the above equation, we get the following equation:

[0111]

[0112] The second-order moment V is estimated according to the following formula t To make a deviation correction:

[0113]

[0114] In the formula, β 2 t is the exponential decay rate of the second-order moment at the current moment.

[0115] The deviation correction process is the same as that of the second-order moment estimation. The following formula is used to estimate the first-order moment M: t To make a deviation correction:

[0116]

[0117] In the formula, β 1 t is the first-order moment exponential decay rate at the current moment.

[0118] In this embodiment, the variable x is defined as the control efficacy matrix, and the update of the variable x is as follows:

[0119]

[0120] In the formula, η∈(0,1] is the learning rate; ε is a very small constant, which can be taken as 1×10 -8 , used to avoid the second-order moment estimate V in the calculation process t When x is zero; t =eff t represents the updated regulatory efficacy matrix, x t-1 =eff t-1 It represents the control efficacy matrix before updating. The above content can be used to realize the iterative update of the control efficacy matrix.

[0121] Since the adjustment amount of the unit roll shifting is very small and the response is slow, it is impossible to solve the control effect of the roll shifting. For the processed data, the adaptive moment estimation algorithm is used to calculate the control effects of Tilt, WRB and IRB respectively, and compared with the control effects provided on site. The identification results are as follows: Figure 6As shown in the figure, the control efficiency obtained by adaptive moment estimation is similar to the trend and value range of the on-site control efficiency curve. The control performance of each mechanism is consistent with empirical knowledge, but slightly different in details.

[0122] Starting from the actual production data of the industrial site, the present invention performs data preprocessing operations such as outlier processing and data matching on the process record data and the shape meter data to establish a data set; performs noise reduction processing on the shape meter data, and uses the adaptive moment estimation algorithm to intelligently identify the control effect of the processed data set. The control ability of the control mechanism is analyzed and clarified, providing more control information for the subsequent shape control calculation and control quantity.

[0123] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A data-driven intelligent identification method for plate shape control efficacy, characterized in that: include: Acquire industrial data in the production process of the rolling mill unit, preprocess the industrial data, and establish a data set based on the preprocessed industrial data, wherein the industrial data includes process record data and flatness meter data; The shape meter data in the data set is subjected to denoising and decomposition and reconstruction, and an adaptive moment estimation algorithm is used to intelligently identify the control efficacy of the denoising and decomposition and reconstruction data set.

2. The data-driven intelligent identification method for plate shape control efficacy according to claim 1 is characterized in that: Preprocessing the industrial data includes: performing outlier processing on the process record data.

3. The data-driven intelligent identification method for plate shape control efficacy according to claim 1 is characterized in that: The datasets established based on preprocessed industrial data include: Based on the exit strip length, the pre-processed process record data and the shape meter data are matched in time series to construct the data set.

4. The data-driven intelligent identification method for plate shape control efficacy according to claim 3 is characterized in that: The process of performing the corresponding matching includes: respectively acquiring a first shape deviation in the process record data and a second shape deviation in the shape meter data, and performing the corresponding matching based on a comparison and verification of the first shape deviation and the second shape deviation.

5. The data-driven intelligent identification method for plate shape control efficacy according to claim 1 is characterized in that: The denoising process of the shape meter data in the data set includes: The wavelet transform algorithm is used to perform noise reduction on the shape meter data, and the formula is: s(n)=f(n)+σe(n) Where s(n) is the actual noisy signal; f(n) is the effective signal; n is the sampling time; σ is the noise intensity; and e(n) is the noise component.

6. The data-driven intelligent identification method for plate shape control efficacy according to claim 5 is characterized in that: Using the wavelet transform algorithm to perform noise reduction on the shape meter data includes: The shape meter data is subjected to multi-layer decomposition of the original signal, and max_level = log2(l data / l filter ) calculate the number of layers of the multi-layer decomposition; after determining the number of decomposition layers, optimize the threshold of the wavelet coefficient using a genetic algorithm; finally, reconstruct the signal using an inverse wavelet transform to obtain the denoised plate shape meter data.

7. The data-driven intelligent identification method for plate shape control efficacy according to claim 1, characterized in that: Decomposing and reconstructing the shape meter data in the data set includes: Decomposing the shape meter data based on Legendre orthogonal polynomials to obtain coefficients λ1, λ2, λ3 and λ4 of each shape basis mode; The shape meter data is reconstructed based on the coefficients λ1, λ2, λ3 and λ4 of each shape basis mode to obtain the decomposed and reconstructed shape meter data.

8. The data-driven intelligent identification method for plate shape control efficacy according to claim 1, characterized in that: Intelligent identification of the control effect of the data set after noise reduction and decomposition and reconstruction using the adaptive moment estimation algorithm includes: Based on the adaptive moment estimation algorithm, the learning rate is automatically updated according to the data characteristics of the data set using the first-order and second-order moment estimates of the gradient, and the first-order and second-order moment estimates are biased, and an updated calculation model for the variable x is defined based on the bias correction of the first-order and second-order moment estimates. The variable x is updated and calculated based on the updated calculation model of the variable x, thereby realizing intelligent identification of the regulation efficacy.

9. The data-driven intelligent identification method for plate shape control efficacy according to claim 8, characterized in that: The update calculation model of the variable x is: In the formula, x t is the x variable at the current moment; t-1 is the x variable at the previous moment; η∈(0,1] is the learning rate; ε is a constant; is the bias correction for the first-order moment estimate; is the bias correction for the second-order moment estimate.

Images